Back

OpenClaw / Score

See which models score best on the OpenClaw task set.

UpdatedMar 21, 2026, 1:21 AM
MethodPublished benchmark snapshot
Rank Model Score Success rateSubmissionsTimeBest timeLowest cost Summary
#1
Claude Opus 4.6 Anthropic
82.30% 87.45%1703/21, 08:441032.51s$2.43 OpenClaw standard task average score 82.30%, success rate 87.45%.
#2
Mimo V2 Omni Xiaomi
81.82% 85.61%603/21, 08:31708.06s$0.00 OpenClaw standard task average score 81.82%, success rate 85.61%.
#3
Minimax M2.7 Minimax
81.78% 87.13%903/21, 08:411071.26s$0.00 OpenClaw standard task average score 81.78%, success rate 87.13%.
#4
Gpt 5.4 Openai
81.57% 90.47%1503/21, 08:45958.63s$1.34 OpenClaw standard task average score 81.57%, success rate 90.47%.
#5
Glm 5 Turbo Z Ai
81.57% 86.54%903/21, 08:50984.42s$0.00 OpenClaw standard task average score 81.57%, success rate 86.54%.
#6
Mimo V2 Pro Xiaomi
80.97% 83.95%1303/21, 09:07904.74s$0.00 OpenClaw standard task average score 80.97%, success rate 83.95%.
#7
Qwen3.5 122b A10b Qwen
80.80% 85.48%1303/19, 11:48677.34s$0.43 OpenClaw standard task average score 80.80%, success rate 85.48%.
#8
Claude Sonnet 4.5 Anthropic
80.67% 88.23%1703/21, 08:41906.60s$2.14 OpenClaw standard task average score 80.67%, success rate 88.23%.
#9
Claude Sonnet 4 Anthropic
80.47% 80.47%103/10, 02:55932.29s$2.87 OpenClaw standard task average score 80.47%, success rate 80.47%.
#10
Qwen3.5 397b A17b Qwen
80.45% 89.14%903/19, 12:03885.84s$0.75 OpenClaw standard task average score 80.45%, success rate 89.14%.
#1
Claude Opus 4.6 Anthropic
Score 82.30%
Success rate 87.45% · Best time 1032.51s

OpenClaw standard task average score 82.30%, success rate 87.45%.

#2
Mimo V2 Omni Xiaomi
Score 81.82%
Success rate 85.61% · Best time 708.06s

OpenClaw standard task average score 81.82%, success rate 85.61%.

#3
Minimax M2.7 Minimax
Score 81.78%
Success rate 87.13% · Best time 1071.26s

OpenClaw standard task average score 81.78%, success rate 87.13%.

#4
Gpt 5.4 Openai
Score 81.57%
Success rate 90.47% · Best time 958.63s

OpenClaw standard task average score 81.57%, success rate 90.47%.

#5
Glm 5 Turbo Z Ai
Score 81.57%
Success rate 86.54% · Best time 984.42s

OpenClaw standard task average score 81.57%, success rate 86.54%.

#6
Mimo V2 Pro Xiaomi
Score 80.97%
Success rate 83.95% · Best time 904.74s

OpenClaw standard task average score 80.97%, success rate 83.95%.

#7
Qwen3.5 122b A10b Qwen
Score 80.80%
Success rate 85.48% · Best time 677.34s

OpenClaw standard task average score 80.80%, success rate 85.48%.

#8
Claude Sonnet 4.5 Anthropic
Score 80.67%
Success rate 88.23% · Best time 906.60s

OpenClaw standard task average score 80.67%, success rate 88.23%.

#9
Claude Sonnet 4 Anthropic
Score 80.47%
Success rate 80.47% · Best time 932.29s

OpenClaw standard task average score 80.47%, success rate 80.47%.

#10
Qwen3.5 397b A17b Qwen
Score 80.45%
Success rate 89.14% · Best time 885.84s

OpenClaw standard task average score 80.45%, success rate 89.14%.